Ü. Doğan; "A survey on Driving Modeling", July 20, 2006, 10:40, G032
  • FENS
  • Ü. Doğan; "A survey on Driving Modeling", July 20, 2006, 10:40, G032

You are here

Faculty of Engineering and Natural Sciences
FENS SEMINARS


A survey on Driving Modeling
Ürün Doğan

Automotive industry and researchers expend considerable effort to develop driver assistance systems. DASs are aware of certain driving situations and support drivers through information, warnings, or even intervention. Many DAS applications are safety oriented, such as lane departure warning systems. Some are comfort oriented, such as automated parking assistants. Still others are designed to alleviate the driver’s attentional load—for example, traffic signal detectors or the already deployed intelligent cruise control systems. Driver models support DAS design in several ways. First, they let researchers analyze system performance in a variety of driving situations. They also help optimize parameter values for certain performance measures such as the degree to which a desired speed is maintained. More sophisticated yet, driver models can form the basis for classifying driver types, such as aggressive versus defensive. DAS applications can use these classifications to activate different parameter value sets that adjust warning thresholds appropriately for different driver types. In the scope of this talk, a survey of driver modeling explained. Driver models that were represented in the literature are classified such as control theoretic models, high level models and driver based models.


Driver Lane Change Motivation Detection


Lane change maneuvers are the most critical maneuvers which a driver has to perform in urban/highway traffic. Nearly one-fourth of the accidents occur due to wrong decision during lane change maneuvers. In order to prevent accidents, driver assistance systems have to be developed, which are able to estimate traffic situations and to predict driver behavior as well as the motivation of the driver to initiate an action. The key element for the context sensitive decision making process is driver modeling. In this respect we propose an approach based on recurrent neural networks, which is capable of modeling driving behavior of humans and to predict driver’s motivation to change lane. In a series of experiments data from human drivers is collected using the simulation environment developed at the Institute of Neuroinformatik. The evaluation of the neural networks showed that the relevant parameters for the prediction of the motivation to change lane are lane curvature, lane offset, derivative of lane offset, steering angle and time to contact. An optimal set of parameters is determined, which predicts the lane change motivation with a false positive rate of approximately zero. In addition the
classification of driving style will be discussed.

Lane change maneuvers are the most critical maneuvers which a driver has to perform in urban/highway traffic. Nearly one-fourth of the accidents occur due to wrong decision during lane change maneuvers. In order to prevent accidents, driver assistance systems have to be developed, which are able to estimate traffic situations and to predict driver behavior as well as the motivation of the driver to initiate an action. The key element for the context sensitive decision making process is driver modeling. In this respect we propose an approach based on recurrent neural networks, which is capable of modeling driving behavior of humans and to predict driver’s motivation to change lane. In a series of experiments data from human drivers is collected using the simulation environment developed at the Institute of Neuroinformatik. The evaluation of the neural networks showed that the relevant parameters for the prediction of the motivation to change lane are lane curvature, lane offset, derivative of lane offset, steering angle and time to contact. An optimal set of parameters is determined, which predicts the lane change motivation with a false positive rate of approximately zero. In addition theclassification of driving style will be discussed.


Ürün Doğan, Research Member of the Institut für Neuroinformatik at the Ruhr-Univeristät Bochum.


July 20, 2006, 10:40, FENS G032